Fundamental Tradeoffs in Distributionally Adversarial Training
Mohammad Mehrabi * 1 Adel Javanmard * 1 Ryan A. Rossi 2 Anup B. Rao 2 Tung Mai 2
Abstract performance, and have been deployed in a multitude of ap-
Adversarial training is among the most effec- plications domains ranging from visual object classification
tive techniques to improve robustness of models to speech recognition, robotics, natural language processing
against adver ...


雷达卡




京公网安备 11010802022788号







